Musée National du Château de Malmaison, Rueil

Anne-Louis Girodet: Ossian Receiving the Ghosts of the French Heroes, circa 1800

In 1811, Jean-Auguste-Dominique Ingres was commissioned to paint an image for the ceiling of Napoleon Bonaparte’s bedroom at the Palazzo del Quirinale in Rome. The former papal palace was being prepared for the French emperor’s visit to the city to assume the title King of Rome and make himself the explicit heir to the Caesars. But Ingres’s painting is not, as might be expected, on a Roman theme. It shows instead the supposedly ancient Gaelic bard Ossian, a blind white-haired figure slumped over his Celtic harp, dreaming of the figures who hover above him. Among them are some of the central figures of Irish mythology, including the warrior and hunter Finn mac Cumaill.

Ingres knew that Napoleon was besotted by the poems of Ossian, which were in fact fabrications woven in the 1760s by the Scottish writer James Macpherson from traditional materials but passed off as rediscoveries of the lost works of a third-century-AD warrior-bard. Napoleon had previously commissioned for his home at the Château de Malmaison an even stranger painting: Anne-Louis Girodet’s bizarre Ossian Receiving the Ghosts of the French Heroes, in which the Gaelic bard welcomes into the afterlife the French generals killed in the revolutionary wars.

It is a peculiar thought that in some parallel universe where Napoleon completed his dominion over Europe, the gods and superhumans of pre-Christian Ireland might have become the official icons of a French-dominated continent. Instead, they remained what they have been for 1,500 years: elusive and angular figures whose very ambiguity has made them supremely adaptable. Paradoxically, these haughty heroes have proven to be most obedient servants, lending themselves at different times to the needs of medieval Irish poets and Victorian theosophists, of Napoleonic propaganda and New Age visions, of W.B. Yeats and J.R.R. Tolkien.

It is a fascinating history and in Mark Williams’s Ireland’s Immortals it has found a magnificent historian. Williams, an Oxford-based scholar of Welsh, Irish, and English medieval literature, is equally at home in the arcana of Old Irish texts and modern English-language writing, and it is this range of erudition that has allowed him to write the first full overview of the long twilight of the Irish gods. Ireland’s Immortals is not just a history of their afterlife—it deserves to be seen as itself a part of that history.

At the heart of this story is a kind of tenderness: the reluctance of one culture to obliterate entirely the one that it had supplanted. To get some idea of the original meaning of those figures like Finn mac Cumaill in Ingres’s painting, we can turn to a book composed around 1220 AD but bringing together many of the older stories of an imagined pre-Christian Ireland. Accalam na Senórach (The Colloquy of the Elders) is a novel-length evocation of the coming of Christianity to Ireland in the fifth century and its rapid replacement of the complex belief systems that Ireland shared with other “Celtic” cultures.1 It is also a moving reflection on memory and forgetting in which we feel the tension between the need on the one hand to banish beliefs that are now heretical and possibly satanic and, on the other, the desire to honor the ancestors who held those beliefs. How, as Christians, could the unknown authors of the book even justify writing down the old pagan stories? Yet how, as Irish people, could they justify the consignment of the culture of their ancestors to oblivion?

Finn mac Cumaill, who is at the center of this particular cycle of stories, seems, from a Christian point of view, quite a dangerous figure. He may originally have been the god of the all-male bands of young aristocratic hunter-warriors, the fían, who roamed Ireland. This institution was anathema to the church—it was nomadic, blasphemously oath-bound, and suspected of being a reservoir of pagan practices: the clergy called the fían “sons of death.” So it is all the more remarkable that the authors of the Accalam, most likely clerics themselves, should choose to preserve some of the lore of these suspected pagans for posterity.

Thus the ingenious narrative device of the Accalam: the last survivors of the fían are allowed to tell their tales to an unimpeachable audience: Saint Patrick himself, who had been born in Roman Britain and became the father of Irish Christianity. Patrick on his missionary travels in the fifth century is blessing an old fort that was once the house of Finn when the last survivors of the fían appear: “His priests…were seized with fear and horror at the sight of these enormous men, the warriors of an earlier age, together with their great dogs.” Patrick sprinkles the giant warriors with holy water and “the demons fled from them in all directions, into the hills and rock-clefts and off to the far reaches of the country.” But it is not just the old pagans who are thus spiritually cleansed—so, more importantly, are the stories they will tell. Patrick, in his generosity, has chosen to allow certain tales, and the gods and heroes they contain, to enter into Irish memory.


What is so moving is the feeling of remembrance being pulled out of the fire of oblivion. The old giants sometimes introduce their tales with phrases like “That story is still clear in my memory,” implying that others are not. Indeed, his two guardian angels tell Patrick that the fían warriors to whom he speaks have already lost fully two thirds of their store of tales “because their memories are faulty.” The angels give him sanction from God himself to keep listening to the stories. This, then, is a rather unbiblical God, one who is gentle enough to grant some of his pagan predecessors the privilege of not being entirely forgotten. He is God the antiquarian. And Patrick, as God’s servant, has a power that the great old warriors, for all their heroic strength, do not possess: writing.

Pre-Christian Irish culture was almost entirely oral and the Accalam is, among other things, a reflection on the precariousness of oral cultures and the wonders of literacy. It is, in the simplest sense, self-consciously literary. The process it describes is that of the translation of the spoken into the written, and of the oral past into a literary future. This act of writing, moreover, has divine sanction. The angels pass on to Patrick God’s instruction to “have these stories written down on poets’ tablets in refined language.” God himself, it seems, wants the Irish deities to be transformed into literature.

And in a sense, as Williams shows with such brilliant persuasiveness, they are nothing but literature. Irish mythology has its own mythology—the idea that it is a clear window on Western Europe’s pagan past, a pristine survival from the Celtic culture that was broken by the Romans and gradually driven to the continent’s western margins. It is an enormously attractive idea—the appeal of the “Celtic” is that it is wild produce, sourced in the ancient forests and on misty crags, not like the farmed stuff of our classical, rational civilization. But Williams demolishes this idea with extraordinary erudition and devastating wit. Happily, though, he shows the Irish gods and heroes to be actually more interesting figures, not static remnants of the deep past but living, infinitely adaptable images that continue to feed the artistic imagination.

When the angels told Patrick that the old heroes had forgotten two thirds of their stories, they were understating the case. The Irish gods are one percent life and 99 percent afterlife. Of the actual religious life of pre-Christian Ireland, its rituals and cults as well as its cosmological and theological beliefs, we know very little. Culturally, the Irish were part of a broader Celtic world. This is not to say that they were themselves Celts: they were descended from the first people who settled the island about 10,000 years ago.

The once-cherished notion of a Celtic invasion of Ireland around three thousand years ago has long since been debunked—there is not a shred of evidence for it. But there is little doubt that Irish culture existed within an Atlantic sphere that encompassed Britain, northern France, and northern Spain. This, though, does not help us very much. As the Patrick of the Accalam understood, things that are not written down disappear and the people who actually believed in the old gods did not write down their own myths. For contemporary evidence, we have only the fragmentary (and often self-serving) impressions left by a few Greek and Roman writers.

A single paragraph in Julius Caesar’s history of his Gallic wars is pored over again and again because it is the nearest thing to an account of Celtic religion—and in it Caesar tells us that the chief god of the Gauls is Mercury, and that after him come Apollo, Mars, Jupiter, and Minerva. It is like a contemporary American going through a gallery of medieval portraits and telling us that they look like Brad Pitt, Angelina Jolie, Arnold Schwarzenegger, Taylor Swift, and Donald Trump. We have some idea of what he is seeing but not much idea of what he is looking at. In any case, we simply do not know whether the Irish gods were precisely the same as those worshiped by the Gauls.2

It seems clear enough that one of the Gaulish gods, Lugdus, the most likely candidate for Caesar’s Mercury-like deity, was widely worshiped in Ireland, though he may have had many local avatars, including the Finn mac Cumaill who ends up on Ingres’s painting for Napoleon’s ceiling and the other great warrior Cú Chulainn, hero of many poems and plays by Yeats. On the other hand, the single most important aspect of the Irish Otherworld is the notion that its denizens dwell inside hollow hills called síde. These were often megalithic tumuli built as sites of worship thousands of years before the emergence of a Celtic culture in Ireland, but they were now understood to be worlds in themselves, much vaster on the inside than on the outside.


The people of the síde, later anglicized as shee and etiolated into Victorian fairies, were dazzlingly beautiful, wealthy, supernaturally powerful, and long-lived or even immortal. The Irish version of Lugdus, called Lug, is explicitly referred to in early written sources as “from the hollow hills.” But this central aspect of Irish belief is not found in Gaul, or even in neighboring Britain. So far as the evidence goes, it seems to be utterly distinctive to Ireland. Ireland’s pre-Christian belief system, then, may well be a mix of Western European cults adapted to local needs and entirely indigenous ideas. Given the highly localized nature of Irish society—it was a patchwork of petty kingdoms—it may well be that practices and beliefs, and even deities, varied considerably across the island.

What is certain is that by the early Middle Ages, when these beliefs are written down in stories that have been heavily filtered through Christianity and literacy, the Irish gods look nothing like the Greco-Roman or Norse pantheons. The síde are not a Celtic Olympus or Valhalla inhabited by relatively stable families of gods with clearly defined attributes. Their inhabitants are closer to superstars and supermodels than to the classical gods: as Williams puts it “they physically resemble us—or would, were we all gorgeous, splendidly dressed young adults in glowing health.” These god-peoples are close enough to mere mortals that they can fall in love with and marry each other—though usually with unhappy consequences. In medieval Gaelic literature, these exquisite deities acquired the collective name Túatha Dé, the god-peoples, which later mutated into the form now in general use, Túatha Dé Danann.3

By the time of the influential pseudo-history The Book of Invasions, written in the eleventh century, the Túatha Dé Danann had been reimagined as a race of people who invaded Ireland in the distant past and routed the indigenous natives, before themselves being defeated by the invading Gaels and relegated to their new kingdoms in the hollow hills. The strangeness of this story, which has the Irish Gaels going to war against and militarily defeating their own gods, is one expression of the difficulty of placing the god-peoples within a Christian worldview. Having rescued them from oblivion, Irish Christian writers have to decide who and what they are. The glory of the old gods, though, is that they do not provide answers to those questions. They refuse to be defined, and thus have to be imagined.

W. B. Yeats
W. B. Yeats; drawing by David Levine

The easy way to think of them, from a Christian perspective, is as satanic demons and sorcerers, and some early Christian writers in Ireland do indeed think of them in this way. But the desire to remember them in writing demands other possibilities. Were they, perhaps, a race of humans with enhanced capacities? Could they be people who had somehow escaped the consequences of the Fall of Adam and Eve? Or were they actual gods who had died out when Jesus saved mankind from its sins? One especially ingenious Irish solution to these ontological questions was the suggestion that the Túatha Dé were in fact “half-fallen” angels who had been expelled from heaven for being neutral in the conflict between God and Satan but, being not quite bad enough for hell, had been allowed to land in Ireland instead. Or, in a particularly daring variation on this theme, perhaps they were in fact good angels who had come to Ireland to help its people before the arrival of Christianity.

The point about these possibilities is that they continued to exist simultaneously, sometimes within the same story. The old gods have a hovering, wavering kind of existence. For ordinary Irish people, they may have retained a vivid presence in oral tales, but we don’t really know how they thought about them. For the literate intellectuals on whose anonymous labors our own knowledge depends, the old gods created a field of ambiguity and uncertainty that invited literary exploration of the slipperiness and insecurity of human existence itself. Indeed, the great attraction of the Irish gods for writers is what they are not. They are not sacred. Actual paganism had almost certainly died out in Ireland by the early eighth century when the last of the druids are lumped in a law tract with “satirists and inferior poets and farters and clowns and bandits and pagans and whores.” The Irish gods have the enormous advantage for writers of not being off-limits. One can say anything one likes about them. As well as being a field of ambiguity, they are also a realm of imaginative freedom.

Many of these old gods probably had deep roots in pre-Christian Ireland: among them are the father figure called the Dagda; Brigit, who is the exemplar of poetry, medicine, and metalwork; the aforementioned Lug; the Morrígan, a goddess of battle who appears as a crow or raven; the sea god Mannanán; the warrior king Núada of the Silver Arm; the beautiful “young lad” Óengus (later called Aengus or Angus); and his mother Bóand, who was the goddess of the River Boyne. Their pure forms are no longer discernible through the fog of oral memory and the thick layers of Christian reinterpretation.

But just as many of the original gods were lost in the transition to Christianity, anonymous medieval writers invented hundreds of others and reinterpreted some of the old gods in the light of Christianity, of classical learning, and of contemporary needs. Brigit, who is now the most popular goddess among contemporary neopagans, was conflated with an early Irish Christian saint of the same name to such an extent that it is impossible to tell whether the attributes of the goddess are back-projections of the saint or vice versa.

What is especially intriguing, though, is that in this process of reinvention, which flourished most fruitfully between the eighth and thirteenth centuries, the Irish poets and prose writers shifted the gods from the realm of nature to that of culture. They probably began as spirits of the sun, moon, sea, and rivers, but in medieval Ireland, they are transformed into images of skill, knowledge, and the arts. They function as prototypes of what Gaelic society called the áes dána, the “people of talent.” In one ninth-century text, a prodigious teenage poet announces his genealogy, beginning with “I am son of Poetry,/Poetry son of Scrutiny,/Scrutiny son of Meditation” and working his way back to “Understanding son of Wisdom,/Wisdom, son of the Three Gods of Skill.” These three gods of skill are glossed in the text as the three sons of the goddess Brigit. Here, as in many other texts, we can see that the old gods are being used to suggest that the artist has a genealogy that is at least the equal of the aristocrat’s. There is thus a lovely symbiosis: the old gods validate the work of the artists and the artists repay the compliment by giving the old gods new life in their vivid tales.

Because the writers are free to invent and because the nature of the gods is so uncertain, these tales have a remarkable variety of tone. On the one hand, for example, there is the great beauty and poignancy of the ninth-century saga The Wooing of Étaín. Midir, a member of the Túatha Dé, falls in love with the mortal Étaín. When Midir takes her home with him, his existing wife, a sorceress, is not best pleased and turns Étaín into a pool of water. But the heat of the fire in the hearth turns Étaín into a purple fly “the size of a man’s head and the most beautiful creature in all the land.”

Eventually, after a thousand years, the fly falls into the cup of another woman who swallows her and ingests her into her womb—a rare instance of same-sex impregnation. Étaín, after her millennium as a fly, is born again in human form, grows up, and marries. She is unaware of her previous existence, but Midir rediscovers her and, after various adventures, the two of them disappear through the skylight in the form of swans.4

On the other hand, the stories can have an earthy irreverence that was deeply uncomfortable to later revivalists. We might expect the Dagda, who seems to be the equivalent of Zeus or Odin, to be approached with awe. But as he appears in one of the most important sagas, The Second Battle of Moytura, written probably in the late ninth century and telling the story of a conflict between the Túatha Dé and another supernatural race, the Fomorians, the Dagda is “naked, with a long penis.” He also engages in one of the strangest sex scenes in medieval literature. The Fomorians torture him by force-feeding him with porridge. He then encounters the beautiful daughter of the Fomorian king who taunts him for his impotence and beats him. She thrusts him waist-deep into the earth, causing the bloated Dagda to lose control of his bowels, which apparently restores his sexual potency:

Then the Dagda got out of the hole, after letting go of the contents of his belly, and the girl had waited for that a long time…. The girl jumped on him and whacked him across the arse, and her curly bush was revealed. At that point the Dagda gained a mistress, and they had sex. The mark remains at Beltraw Strand where they coupled.

It is not surprising that this part of the story was entirely excised when Augusta, Lady Gregory, Yeats’s great ally and collaborator, published the most elegant revival of the Túatha Dé tales as Gods and Fighting Men in 1904. (The fly into which Étaín was transformed became a more decorous butterfly.) But Gregory was in her own way being consistent with the tradition—the gods owed their continued existence to their adaptability. In the Celtic Revival led by Yeats and Gregory in the late nineteenth and early twentieth centuries, they were repurposed to meet a variety of demands. For an Anglo-Irish landowning class that was gradually being displaced by the rise of the Catholic peasantry and middle class, the Túatha Dé provided a fantasy of a dazzling elite surviving its own defeat. George Russell thought the old Irish gods were avatars of the Hindu deities and imagined that a Celtic theosophy would become the new national religion. These fantasies transcended their Irish settings, feeding into a Celtic revival in Scotland, into Californian New Age hippiedom, into Tolkien’s Lord of the Rings, where the elves are the Túatha Dé in thin disguise, and into contemporary beliefs ranging from Wiccan cults to feminist and “Celtic” theology.

For Yeats, the Irish gods initially promised a little more than they delivered. He put enormous energy into trying to conjure them in visions and magical rituals but they declined to make satisfactory appearances. They are, as Yeats came to understand, purely imaginary beings. He came to focus on one of them, Óengus/Aengus, as the embodiment of the poetic imagination itself. In the most beautiful revivification of the dead gods, he picked up on an enigmatic eighth-century saga, The Dream of Óengus, in which the young god languishes after nightly visions of a beautiful but unattainable woman. With the great bravado that only he could muster, Yeats stopped trying to evoke the god and instead simply became him. The “I” of “The Song of Wandering Aengus” is both Yeats himself and the golden “young lad” of the Túatha Dé:

I went out to the hazel wood,
Because a fire was in my head,
And cut and peeled a hazel wand,
And hooked a berry to a thread….

In a transformation worthy of the old stories themselves, he catches a fish that turns into a young woman who then disappears. His pursuit of her will stretch into infinity while he imagines how they will together

walk among long dappled grass,
And pluck till time and times are done
The silver apples of the moon,
The golden apples of the sun.

When the angels asked Patrick to “have these stories written down on poets’ tablets in refined language,” they must have had in mind Yeats’s revival of the old god as a luminous presence that is always wandering, never quite captured.